From mboxrd@z Thu Jan 1 00:00:00 1970 From: Akira Hayakawa Subject: Re: [dm-devel] [RFC] dm-lc target Date: Tue, 30 Jul 2013 22:02:51 +0900 Message-ID: <51F7B97B.3010704@gmail.com> References: <51F48531.6040104@gmail.com> <20130729143916.GA20119@debian> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20130729143916.GA20119@debian> Sender: linux-kernel-owner@vger.kernel.org To: device-mapper development Cc: sunnikrishnan@stec-inc.com, kmo@daterainc.com, linux-kernel@vger.kernel.org List-Id: dm-devel.ids Thanks for your comment, Joe. Yes, dm-lc focuses solely on optimizing writes. As you know, writing to RAID storage is slower than reading because of so-called RAID penalty. So, only improving writes may rescue most of the actual storage systems suffering from heavy I/Os. That's why I cares writes more than reads. > The next step is to get some third party users to post their > experiences with it, and get some feedback on the code itself from > Mike Snitzer. Yes, I want to make my code better. > If I get time I'll integrate it with my test suite. Thanks. That seems to be a nice test suite. I am currently testing dm-lc by make -j N and then make test with ruby source code, but I think the test should be severer. Actually, I have tried to implement general test tool for block device. But I gave up. https://github.com/akiradeveloper/iocheck Your test suite is like one I have been pursuing. I am really happy if you integrate dm-lc to your test suite. Akira On 7/29/13 11:39 PM, Joe Thornber wrote: > On Sun, Jul 28, 2013 at 11:42:57AM +0900, Akira Hayakawa wrote: >> I wrote an introduction slide which is available in URL below: >> https://github.com/akiradeveloper/dm-lc/blob/develop/what-is-dm-lc.pdf >> access the page and click "view raw". > > This is nicely orthogonal to dm-cache. > > dm-cache tries to move a working set of fixed size blocks into a cache > to optimise both read and write. dm-lc focusses solely on optimising > bursty writes and uses a journal based structure to accomodate > variable sized blocks. > > The next step is to get some third party users to post their > experiences with it, and get some feedback on the code itself from > Mike Snitzer. > > If I get time I'll integrate it with my test suite. > > https://github.com/jthornber/device-mapper-test-suite > > > - Joe > > -- > dm-devel mailing list > dm-devel@redhat.com > https://www.redhat.com/mailman/listinfo/dm-devel >